Very nice read explaining why if your variables that you want to correlate are measured with error, a simple correlation underestimates the true relationship, and you need to use methods that account for that error, rather than simply correlating summary statistics like line means.
add a skeleton here at some point
3 months ago